Analysis Paralysis: What a Fighter Pilot Discovered About Making Decisions Faster

During the Korean War, American F-86 Sabre pilots were dominating enemy MiG-15s at a ratio that didn't make sense. The Air Force initially claimed ten to one. Later analysis revised the number downward, but even conservative estimates put American pilots ahead by a wide margin. The MiG-15 was, on paper, the superior aircraft. It climbed faster, turned tighter at high altitude, and carried heavier armament. Soviet engineers had built a plane that should have held its own in every dogfight over the Korean Peninsula. Instead, American pilots kept winning.

The U.S. Air Force studied the discrepancy for years. They tested the aerodynamics, compared the specs, interviewed the pilots. Nothing in the hardware explained the advantage. Then a military strategist named John Boyd, who had flown F-86s in Korea and later became one of the most influential tactical thinkers in modern military history, proposed an answer that had nothing to do with the planes and everything to do with the pilots' brains. Analysis paralysis, the state where overthinking a decision makes it impossible to act, is the default failure mode of every intelligent person facing uncertainty. The antidote isn't better analysis. It's a faster cycle between observation, decision, and action. Boyd realized the F-86 had two design features the MiG lacked: a hydraulic flight control system that responded faster to pilot inputs, and a bubble canopy that gave pilots a wider field of vision. Neither feature made the plane faster or more lethal. But both made the pilot faster. The American pilots could see what was happening sooner and respond to it sooner. They were cycling through decisions at a speed the MiG pilots couldn't match.

Over the following decades, Boyd formalized what he'd observed into a framework he called the OODA loop: Observe, Orient, Decide, Act. The core insight was counterintuitive. Victory didn't go to the side with more information or better analysis. It went to the side that could complete the decision cycle faster. If you could get inside your opponent's decision loop, making your next move before they'd finished processing your last one, you won regardless of who had the better position on paper.

The principle applies far beyond aerial combat. Boardrooms, product launches, hiring, pricing: the same pattern holds everywhere decisions carry stakes. The company that makes good decisions quickly beats the company that makes perfect decisions slowly. And the most common way intelligent founders lose isn't by making the wrong call. It's by making no call at all.

What Is Analysis Paralysis?

A Columbia University psychologist named Sheena Iyengar set up a jam-tasting booth at Draeger's, an upscale supermarket in Menlo Park, California. Draeger's was the kind of place that stocked 250 varieties of mustard, 500 types of produce, and roughly 300 varieties of jam. The perfect environment for the experiment Iyengar had in mind.

She alternated the booth between two displays. One offered 24 varieties of Wilkin & Sons jam. The other offered 6. Shoppers could sample as many as they wanted, and each received a coupon for a dollar off any jam purchase.

The large display drew more traffic. Sixty percent of passing shoppers stopped at the 24-jam booth, compared to 40 percent at the 6-jam booth. More variety pulled more attention. But at the register, the pattern reversed. Of the shoppers who stopped at the 6-jam display, 30 percent bought a jar. Of the shoppers who stopped at the 24-jam display, 3 percent bought a jar. A tenfold difference in conversion, driven entirely by the number of options.

Iyengar published the results in 2000 in the Journal of Personality and Social Psychology. The paper became one of the most cited studies in behavioral economics. It also became one of the most contested. In 2010, a meta-analysis by Benjamin Scheibehenne examined fifty experiments on choice overload and found the average effect was near zero. Sometimes more choice increased purchases. Sometimes it decreased them. Scheibehenne himself tried to replicate the jam study for his dissertation and failed.

So is choice overload real or not? The answer that survived the replication era is more useful than the original headline. Choice overload is real, but it doesn't happen every time you add an option to a menu. It happens under specific conditions, the same conditions that make too many product tiers backfire: when the options are complex, when no option is clearly dominant, when the chooser doesn't have pre-existing preferences, and when the decision carries consequences. Those conditions describe almost every strategic decision a founder makes. You're choosing between options that all look plausible, none of which is obviously best, in a domain where you don't yet have enough experience to have strong priors, and where the stakes are real. The jam study might not replicate at every grocery store. But the boardroom version replicates every day.

Herbert Simon, who won the Nobel Prize in Economics in 1978, had named the underlying problem two decades before Iyengar's jam booth existed. He called it bounded rationality: the brain cannot evaluate all alternatives because it doesn't have infinite time, infinite information, or the processing power to run every comparison. The rational response to bounded rationality, Simon argued, was not to try harder. It was to stop trying to optimize and start trying to satisfice, a word he invented by combining "satisfy" and "suffice." A satisficer sets a threshold ("good enough for my purposes") and takes the first option that clears it. A maximizer evaluates every available option searching for the best one. Simon's research showed that maximizers spend more time, experience more regret, and often end up worse off than satisficers who decided faster with less information.

What Happens in Your Brain When You Can't Decide?

In 2018, a team led by Elena Reutskaja published a study in Nature Human Behaviour that put the choice overload question inside an fMRI scanner for the first time. Participants chose landscape images from sets of 6, 12, or 24 options while researchers watched their brains work.

Two regions told the story. The dorsal striatum, which tracks motivation and anticipated reward, and the anterior cingulate cortex, which weighs costs against benefits. Both regions followed an inverted U-shaped curve. Activity peaked at 12 options, the set participants rated as feeling like "the right amount." At 6 options, both regions were less active: not enough variety to feel like a real choice. At 24 options, both regions dropped again, but for a different reason. The cognitive cost of evaluating all those alternatives had exceeded the brain's anticipated benefit of choosing well.

The interesting part came from a control condition. When participants were told to just browse rather than choose, the inverted-U pattern vanished. Looking at 24 options without having to pick one didn't overload anything. The bottleneck wasn't perception. It was the act of deciding itself. When the brain knows a commitment is required, it tries to compute the value of every alternative against every other alternative. Comparisons grow exponentially with the number of options. At some point, the computation stalls. No amount of additional processing time changes that.

From the inside, analysis paralysis feels deceptively productive. The brain keeps searching for the comparison that will make the answer obvious, and the answer never becomes obvious, so the search never terminates. You experience it as a feeling of being stuck, of needing just a little more data, of not being quite ready to commit. The feeling isn't a signal that you need more information. It's a signal that the computation has exceeded your processing capacity and will never converge no matter how long you wait.

How Do the Best Decision-Makers Move Faster?

Jeff Bezos described the solution in his 2015 letter to Amazon shareholders. He sorted all decisions into two categories. Type 1 decisions are one-way doors: irreversible, high-consequence, and worth deliberating carefully. Type 2 decisions are two-way doors: reversible, lower-consequence, and worth making quickly because you can always walk back through.

The problem Bezos identified wasn't that companies made bad decisions. It was that companies treated two-way doors like one-way doors. "As organizations get larger," he wrote, "there seems to be a tendency to use the heavy-weight Type 1 decision-making process on most decisions, including many Type 2 decisions. The end result of this is slowness, unthoughtful risk aversion, failure to experiment sufficiently, and consequently diminished invention."

In a follow-up letter the next year, Bezos proposed a threshold: "Most decisions should probably be made with somewhere around 70% of the information you wish you had." Below that, you're guessing. But if you wait for 90%, you've almost certainly waited too long. Colin Powell reportedly operated on a similar rule: never decide with less than 40% of the information, but always decide before you reach 70%, because by then events are already deciding for you.

McKinsey surveyed over 1,200 executives in 2019. Organizations that made decisions quickly were twice as likely to also make high-quality decisions. Not a trade-off. A correlation. Respondents at winning organizations were twice as likely to report financial returns of at least 20 percent from their recent decisions. Deliberation, past a point, doesn't improve the outcome. It degrades it.

The same research found that managers spend 37 percent of their time making decisions, and more than half of that time is spent ineffectively. The time isn't producing better outcomes. It's producing more anxiety about the same set of options, which is the jam booth all over again, running in a loop. It's the same reason decision fatigue compounds across the day: the brain's evaluation circuits have a finite budget, and deliberation draws it down whether or not it produces a result.

The Two-Door Test

Most of the decisions that paralyze founders are two-way doors disguised as one-way doors. Which landing page headline to test. Which pricing tier to launch with. Whether to add a feature before launch or after. The market segment question. The contractor-versus-DIY question that's been open for six weeks.

None of these are permanent. All of them are reversible.

Every one of them, if delayed long enough, costs more than the wrong answer would have.

The Two-Door Test is a three-second diagnostic. When you catch yourself stalled on a decision, ask one question: "If this turns out to be wrong, can I reverse it or adjust within a few weeks without catastrophic cost?" If the answer is yes, you're standing in front of a two-way door. Walk through it. The information you get from acting will be better than the information you get from thinking, because the information from acting is real and the information from thinking is simulated.

The instinct to over-analyze feels like diligence. Like responsibility. Like the thing a smart person would do. Which is what makes it dangerous, in the same way that willpower feels like a muscle you can push harder when the real bottleneck is environmental design. Your brain rewards the feeling of thoroughness the same way it rewards the feeling of certainty. You can spend three weeks comparing project management tools, reading reviews, watching demos, building comparison spreadsheets. Or you can pick one, use it for a week, and know more than three weeks of research could have told you. One approach feels rigorous. The other produces knowledge.

One-way doors deserve real deliberation. Signing a five-year lease. Taking on a co-founder. Choosing whether to raise venture capital or bootstrap. These reshape the entire landscape and can't be unwound cheaply. Slow down, seek dissenting opinions, stress-test your assumptions the way high-stakes decision-makers do under pressure. But most of what stalls founders never belonged in this category. It landed there because the brain's prediction system couldn't compute a clear winner, and in the absence of a clear winner, the brain's default is to keep computing rather than to act.

Boyd's insight from the Korean War applies directly. A pilot who analyzed the situation 10 percent better but responded two seconds slower lost the engagement. The one who acted on a good-enough read and adjusted in real time owned the sky. Your market works the same way. The information environment changes fast enough that the analysis you're perfecting right now will be partially obsolete by the time you finish it.